ANALYSIS OF FOOD PRODUCTION AND SUPPLY IN AFRICA BY OLUWATOMISIN AROKODARE¶

PROJECT OVERVIEW¶

Food production and supply play an important role in providing nutrition and promoting economic growth in Africa.Food security includes the availability, usability, stability, and affordability of food for both people and communities.Human well-being, societal stability, poverty reduction, and sustainable development all rely on food security.

In this project, I will be analyzing two different datasets namely food production and food supply datasets.The dataset was sourced from Food and Agriculture Organization of United Nations website.The two datasets have information about food production and food supply in 45 African countries between (2004 - 2013) respectively.The Population and Country code dataset was sourced from Kaggle.

Dataset Features¶

Food Production Dataset:

1.Country: The name of the country.

2.Item: The name of the produce from each country.

3.Year: The year of production.

4.Value: The quantity of each item, measured in kilotonnes (kt).

Food Supply Dataset:

1.Country: The name of the country.

2.Year: The year of supply.

3.Value: The quantity supplied, measured in kilocalories per person per day (kcal/person/day).

PROJECT GOAL¶

The goal of of this project is to analyze these datasets, and gain insights on African food shortage problems, and producing visuals that can help in making appropriate decisions and ensuring food security in Africa.

In [1]:
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
from matplotlib import patheffects
import folium
import geopandas as gpd
import os
import plotly.express as px
import plotly.graph_objects as go
import plotly.offline as pyo
from plotly.subplots import make_subplots
world = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))

%matplotlib inline
import datetime as dt
import warnings
warnings.filterwarnings('ignore')

C:\Users\aroko\AppData\Local\Temp\ipykernel_14748\4270348101.py:13: FutureWarning:

The geopandas.dataset module is deprecated and will be removed in GeoPandas 1.0. You can get the original 'naturalearth_lowres' data from https://www.naturalearthdata.com/downloads/110m-cultural-vectors/.

DATA GATHERING- FOOD PRODUCTION DATASET¶

In [2]:
# loading the food production dataset into a pandas dataframe
af_prod= pd.read_csv('Africa Food Production (2004 - 2013).csv')
af_prod
Out[2]:
Country Item Year Value
0 Algeria Wheat and products 2004 2731
1 Algeria Wheat and products 2005 2415
2 Algeria Wheat and products 2006 2688
3 Algeria Wheat and products 2007 2319
4 Algeria Wheat and products 2008 1111
... ... ... ... ...
23105 Zimbabwe Crustaceans 2009 0
23106 Zimbabwe Crustaceans 2010 0
23107 Zimbabwe Crustaceans 2011 0
23108 Zimbabwe Crustaceans 2012 0
23109 Zimbabwe Crustaceans 2013 0

23110 rows × 4 columns

ASSESSING THE DATASET¶

In [3]:
### Checking information about foodproduction dataset 
af_prod.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 23110 entries, 0 to 23109
Data columns (total 4 columns):
 #   Column   Non-Null Count  Dtype 
---  ------   --------------  ----- 
 0   Country  23110 non-null  object
 1   Item     23110 non-null  object
 2   Year     23110 non-null  int64 
 3   Value    23110 non-null  int64 
dtypes: int64(2), object(2)
memory usage: 722.3+ KB
In [4]:
### Checking the data types of the columns
af_prod.dtypes
Out[4]:
Country    object
Item       object
Year        int64
Value       int64
dtype: object
In [5]:
### Statistical description of food production dataset
af_prod.describe()
Out[5]:
Year Value
count 23110.000000 23110.000000
mean 2008.498269 327.785201
std 2.871740 1607.940343
min 2004.000000 0.000000
25% 2006.000000 3.000000
50% 2008.000000 18.000000
75% 2011.000000 108.000000
max 2013.000000 54000.000000
In [6]:
### checking for missing values in the dataframe
af_prod.isnull().sum()
Out[6]:
Country    0
Item       0
Year       0
Value      0
dtype: int64
In [7]:
### checking the number of unique values in each food production columns
af_prod.nunique()
Out[7]:
Country      45
Item         94
Year         10
Value      2134
dtype: int64

OBSERVATION:There are 45 countries, 94 Food Items and Ten(10) Years in the Food Production Data.

In [8]:
### checking for duplicated rows in the datafarame
af_prod.duplicated().sum()
Out[8]:
0

DATA CLEANING¶

In [9]:
af_prod.head(5)
Out[9]:
Country Item Year Value
0 Algeria Wheat and products 2004 2731
1 Algeria Wheat and products 2005 2415
2 Algeria Wheat and products 2006 2688
3 Algeria Wheat and products 2007 2319
4 Algeria Wheat and products 2008 1111
In [10]:
### creating a unit column containing the symbol of kilotone, which is the unit of the numbers in the Value column
af_prod['Unit'] = 'kt'
In [11]:
af_prod.head(10)
Out[11]:
Country Item Year Value Unit
0 Algeria Wheat and products 2004 2731 kt
1 Algeria Wheat and products 2005 2415 kt
2 Algeria Wheat and products 2006 2688 kt
3 Algeria Wheat and products 2007 2319 kt
4 Algeria Wheat and products 2008 1111 kt
5 Algeria Wheat and products 2009 2953 kt
6 Algeria Wheat and products 2010 2605 kt
7 Algeria Wheat and products 2011 2555 kt
8 Algeria Wheat and products 2012 3432 kt
9 Algeria Wheat and products 2013 3299 kt
In [12]:
### converting the Year column to datetime
af_prod['Year'] = pd.to_datetime(af_prod['Year'], format="%Y").dt.year
In [13]:
### Groupby Country and year and show the total value for each product for different years
af_prod_1 = af_prod.groupby(['Country', 'Year'])
af_prod_1.first()
Out[13]:
Item Value Unit
Country Year
Algeria 2004 Wheat and products 2731 kt
2005 Wheat and products 2415 kt
2006 Wheat and products 2688 kt
2007 Wheat and products 2319 kt
2008 Wheat and products 1111 kt
... ... ... ... ...
Zimbabwe 2009 Wheat and products 12 kt
2010 Wheat and products 18 kt
2011 Wheat and products 23 kt
2012 Wheat and products 20 kt
2013 Wheat and products 25 kt

450 rows × 3 columns

In [14]:
### Initialize the Total sum of Food produced by each year by (Grouping by Country, Year and Values).
af_prod.groupby(['Country','Year'])['Value'].sum()
Out[14]:
Country   Year
Algeria   2004    15536
          2005    15667
          2006    16417
          2007    14763
          2008    13841
                  ...  
Zimbabwe  2009     5754
          2010     6777
          2011     7551
          2012     8173
          2013     7914
Name: Value, Length: 450, dtype: int64

DATA GATHERING- FOOD SUPPLY DATASET¶

In [15]:
# loading the food supply dataset into a pandas dataframe
af_supply = pd.read_csv('Africa Food Supply (2004 - 2013).csv')
af_supply
Out[15]:
Country Year Value
0 Algeria 2004 2987
1 Algeria 2005 2958
2 Algeria 2006 3047
3 Algeria 2007 3041
4 Algeria 2008 3048
... ... ... ...
445 Zimbabwe 2009 2147
446 Zimbabwe 2010 2168
447 Zimbabwe 2011 2200
448 Zimbabwe 2012 2197
449 Zimbabwe 2013 2110

450 rows × 3 columns

ASSESSING THE DATASET¶

In [16]:
### Checking information about food supply dataset 
af_supply.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 450 entries, 0 to 449
Data columns (total 3 columns):
 #   Column   Non-Null Count  Dtype 
---  ------   --------------  ----- 
 0   Country  450 non-null    object
 1   Year     450 non-null    int64 
 2   Value    450 non-null    int64 
dtypes: int64(2), object(1)
memory usage: 10.7+ KB
In [17]:
### Checking the data types of the columns
af_supply.dtypes
Out[17]:
Country    object
Year        int64
Value       int64
dtype: object
In [18]:
### Statistical description of food supply dataset
af_supply.describe()
Out[18]:
Year Value
count 450.000000 450.000000
mean 2008.500000 2469.775556
std 2.875478 379.140143
min 2004.000000 1781.000000
25% 2006.000000 2174.000000
50% 2008.500000 2376.000000
75% 2011.000000 2681.750000
max 2013.000000 3561.000000
In [19]:
### checking for missing values in the dataframe
af_supply.isnull().sum()
Out[19]:
Country    0
Year       0
Value      0
dtype: int64
In [20]:
### checking for duplicated rows in the datafarame
af_supply.duplicated().sum()
Out[20]:
0
In [21]:
### checking the number of unique values in each food production columns
af_supply.nunique()
Out[21]:
Country     45
Year        10
Value      373
dtype: int64

OBSERVATION:There are 45 countries and Ten(10) Years in the Food Supply Data.

DATA CLEANING¶

In [22]:
af_supply.head(5)
Out[22]:
Country Year Value
0 Algeria 2004 2987
1 Algeria 2005 2958
2 Algeria 2006 3047
3 Algeria 2007 3041
4 Algeria 2008 3048
In [23]:
###Creating a unit column containing the symbol of kilotone, which is the unit of the numbers in the Value column
af_supply['Unit'] = 'kcal/person/day'
In [24]:
af_supply.head(10)
Out[24]:
Country Year Value Unit
0 Algeria 2004 2987 kcal/person/day
1 Algeria 2005 2958 kcal/person/day
2 Algeria 2006 3047 kcal/person/day
3 Algeria 2007 3041 kcal/person/day
4 Algeria 2008 3048 kcal/person/day
5 Algeria 2009 3110 kcal/person/day
6 Algeria 2010 3142 kcal/person/day
7 Algeria 2011 3217 kcal/person/day
8 Algeria 2012 3272 kcal/person/day
9 Algeria 2013 3296 kcal/person/day
In [25]:
### converting the Year column to datetime
af_supply['Year'] = pd.to_datetime(af_supply['Year'], format="%Y").dt.year
In [26]:
### Groupby Country and year and show the total value for each product for different years
af_supply_1 = af_supply.groupby(['Country', 'Year'])
af_supply_1.first()
Out[26]:
Value Unit
Country Year
Algeria 2004 2987 kcal/person/day
2005 2958 kcal/person/day
2006 3047 kcal/person/day
2007 3041 kcal/person/day
2008 3048 kcal/person/day
... ... ... ...
Zimbabwe 2009 2147 kcal/person/day
2010 2168 kcal/person/day
2011 2200 kcal/person/day
2012 2197 kcal/person/day
2013 2110 kcal/person/day

450 rows × 2 columns

In [27]:
### Initialize the Total sum of Food supply by each year by (Grouping by Country, Year and Values).
af_supply.groupby(['Country','Year'])['Value'].sum()
Out[27]:
Country   Year
Algeria   2004    2987
          2005    2958
          2006    3047
          2007    3041
          2008    3048
                  ... 
Zimbabwe  2009    2147
          2010    2168
          2011    2200
          2012    2197
          2013    2110
Name: Value, Length: 450, dtype: int64

EXPLANATORY ANALYSIS¶

QUESTION 1.What is the Trend in food production over the years.¶

In [28]:
# visualizing trends in food production using line plot
plt.figure(figsize=[17,8])

sns.lineplot(data=af_prod, x="Year", y="Value", marker='o', markersize=12, ci=None, color='y')
plt.title("Food Production trends in Africa (2004 -2013)", fontsize=15);
plt.ylabel('Value (Kt)', fontsize=13);
plt.xlabel('Year', fontsize=13);
plt.show()
In [29]:
# visualizing trends in food production(2004-2013)
fig, ax = plt.subplots(figsize=(10, 6))

ax.bar(af_prod['Year'], af_prod['Value'], color='y')
ax.set_title('Food Production Trends in Africa (2004 - 2013)', fontsize=15)
ax.set_xlabel('Year', fontsize=12)
ax.set_ylabel('Value (Kt)', fontsize=12)

plt.show()

Observation: Food production in Africa increases yearly. However, apart from the drop in 2007, there was a progress increase in food production from 2008 to 2013.

QUESTION 2. What is the Trend in food supply over the years?¶

In [30]:
# visualizing trends in food supply(2004-2013)
fig, ax = plt.subplots(figsize=(10, 6))

ax.bar(af_supply['Year'], af_supply['Value'], color='g')
ax.set_title('Food supply Trends in Africa (2004 - 2013)', fontsize=15)
ax.set_xlabel('Year', fontsize=12)
ax.set_ylabel('Value (Kt)', fontsize=12)

plt.show()
In [31]:
# Visualizing trends in food supply using line plot
plt.figure(figsize=[17,8])

sns.lineplot(data=af_supply, x="Year", y="Value", marker='o', markersize=12, ci=None, color='g')
plt.title("Food Supply trends in Africa (2004 -2013)", fontsize=15);
plt.ylabel('Value (Kcal/Person Day)', fontsize=14);
plt.xlabel('Year', fontsize=14);
plt.show()

Observation: There is progressive increase in food supply in African countries from 2004 to 2013.

QUESTION 3. Which countries belong to theTop and Bottom 3 with the highest and Least volume of food production (2004-2013)?¶

In [32]:
# Statistics showing the top countries that produced food the most.
af_prod.groupby(['Country'])['Value'].sum().sort_values(ascending=False)
Out[32]:
Country
Nigeria                        1628030
Egypt                           877498
South Africa                    579592
Ethiopia                        350693
United Republic of Tanzania     332802
Sudan                           301584
Ghana                           295004
Uganda                          278421
Morocco                         271821
Kenya                           256872
Cameroon                        197670
Angola                          195499
Algeria                         188650
Cote d'Ivoire                   182486
Mozambique                      157584
Malawi                          153717
Madagascar                      133071
Burkina Faso                    100129
Rwanda                           95174
Tunisia                          92713
Mali                             92292
Benin                            88072
Niger                            86808
Zambia                           82701
Zimbabwe                         73316
Guinea                           66848
Swaziland                        62665
Mauritius                        51785
Senegal                          49414
Sierra Leone                     47737
Chad                             45078
Togo                             32155
Congo                            25791
Central African Republic         24921
Liberia                          14657
Namibia                          14197
Gabon                            13566
Mauritania                       10039
Guinea-Bissau                     7296
Botswana                          4808
Gambia                            4699
Lesotho                           3785
Cabo Verde                        1658
Sao Tome and Principe             1152
Djibouti                           666
Name: Value, dtype: int64
In [33]:
### Visualizing the Volume of food production per country(2004-2013)
plt.figure(figsize=[12,12])
fig = sns.barplot(data=af_prod,
                  x="Country",
                  y="Value")

fig.set_title('Volume of Food Produced by African Countries (2004 - 2013)', fontsize=15)
fig.set_xlabel('Country', fontsize=12)
fig.set_ylabel('Value (kt)', fontsize=12)

fig.set_xticklabels(fig.get_xticklabels(), rotation=90)
plt.tight_layout()

plt.show()

Observation: The TOP 3 African countries with the highest volume of food production within (2004-2013) are Nigeria, Egypt, & South Africa.

However,the BOTTOM 3 African countries Cabo Verde, Sao Tome & Principe and Djibouti produced the least volume of food.

QUESTION 4. Which Countries belong to the Top and Bottom 3 with the highest and Least volume of food Supply (2004-2013)?¶

In [34]:
# Statistics showing the top countries that produced food the most.
af_supply.groupby(['Country'])['Value'].sum().sort_values(ascending=False)
Out[34]:
Country
Egypt                          34580
Tunisia                        33055
Morocco                        32967
Algeria                        31118
Mauritius                      30543
South Africa                   29629
Ghana                          29180
Cote d'Ivoire                  27666
Mali                           27502
Mauritania                     27443
Gabon                          27299
Nigeria                        26988
Burkina Faso                   26072
Gambia                         25695
Lesotho                        25588
Cabo Verde                     25514
Benin                          25378
Guinea                         25180
Niger                          25024
Cameroon                       24603
Sao Tome and Principe          24462
Djibouti                       24165
Senegal                        23784
Togo                           23339
Sudan                          23238
Swaziland                      23171
Guinea-Bissau                  22963
Malawi                         22925
Angola                         22556
Sierra Leone                   22291
Botswana                       22263
Uganda                         22205
Liberia                        21827
Mozambique                     21702
Namibia                        21602
United Republic of Tanzania    21550
Congo                          21530
Kenya                          21453
Rwanda                         21309
Zimbabwe                       21209
Central African Republic       20719
Madagascar                     20608
Chad                           20511
Ethiopia                       20292
Zambia                         18701
Name: Value, dtype: int64
In [35]:
# Visualizing the volume of food supply per country(2004-2013)
plt.figure(figsize=[12,15])
fig = sns.barplot(data=af_supply,
                  x="Country",
                  y="Value",
                  palette="viridis")

fig.set_title('Volume of Food Supplied by African Countries (2004 - 2013)', fontsize=15)
fig.set_xlabel('Country', fontsize=12)
fig.set_ylabel('Value (kcal/person/day)', fontsize=12)

fig.set_xticklabels(fig.get_xticklabels(), rotation=90)
plt.tight_layout()

plt.show()

Observation: The TOP 3 African countries with the highest volume of food supply within (2004-2013) are Egypt, Tunisia, & Morocco However,the BOTTOM 3 African countries Chad, Ethiopia and Zambia supplied the least volume of food.

QUESTION 5. What Food items was produced the most by African countries within (2004-2013)¶

In [36]:
# Dataframe for the most produced food items
mp_food = pd.DataFrame(af_prod.groupby('Item')['Value'].sum().sort_values(ascending = False)).reset_index()
mp_food
Out[36]:
Item Value
0 Cassava and products 1118948
1 Sugar cane 887008
2 Maize and products 574948
3 Yams 511523
4 Vegetables, Other 410403
... ... ...
89 Cloves 249
90 Molluscs, Other 214
91 Pepper 199
92 Fish, Liver Oil 28
93 Aquatic Animals, Others 13

94 rows × 2 columns

In [37]:
mp_food = pd.DataFrame(af_prod.groupby('Item')['Value'].sum().sort_values(ascending=False)).reset_index()[:5]
plt.figure(figsize=[12, 7])
ax = sns.barplot(y=mp_food.Value, x=mp_food.Item, data=mp_food, color='g')

for i, bar in enumerate(ax.patches):
    count = bar.get_height()

## Labeling each bar
    ax.annotate(f'{int(count)}',
                xy=(bar.get_x() + bar.get_width() / 2, bar.get_height()),
                ha='center', va='bottom', fontsize=10, color='black', fontweight='bold')

plt.gca().set_title("Top 5 Most Produced Food Item in Africa (2004 - 2013)")
plt.ylabel('Value (Kiloton)')
plt.xlabel('Item')

plt.show()

Observation: **The Top 5 most produced food in Africa between 2004 and 2013 are Cassava, Sugar Cane, Maize and products, Yam, and Vegetables.

QUESTION 6. What Food items was least produced the most by African countries within (2004-2013)?¶

In [38]:
# Dataframe for the most produced food items
mp_food = pd.DataFrame(af_prod.groupby('Item')['Value'].sum().sort_values(ascending = True)).reset_index()
mp_food
Out[38]:
Item Value
0 Aquatic Animals, Others 13
1 Fish, Liver Oil 28
2 Pepper 199
3 Molluscs, Other 214
4 Cloves 249
... ... ...
89 Vegetables, Other 410403
90 Yams 511523
91 Maize and products 574948
92 Sugar cane 887008
93 Cassava and products 1118948

94 rows × 2 columns

In [39]:
plt.figure(figsize=[12, 7])
mp_food = pd.DataFrame(af_prod.groupby('Item')['Value'].sum().sort_values(ascending=True)).reset_index()[:5]

ax = sns.barplot(y=mp_food.Value, x=mp_food.Item, data=mp_food, color='g')

for i, bar in enumerate(ax.patches):
    count = bar.get_height()

## Labeling each bar
    ax.annotate(f'{int(count)}',
                xy=(bar.get_x() + bar.get_width() / 2, bar.get_height()),
                ha='center', va='bottom', fontsize=10, color='black', fontweight='bold')

plt.gca().set_title("Least 5 Most Produced Food Item in Africa (2004 - 2013)")
plt.ylabel('Value (Kiloton)')
plt.xlabel('Item')


plt.show()

Observation: **The Least 5 most produced food in Africa between 2004 and 2013 are Cloves, Molluscs,Pepper, Fish,Liver oil and Aquatic Animals.

QUESTION 7. What are the most popular Food Items that the Top 3 producing Africa Countries Produced ?¶

In [40]:
#Let's put the spotlight on the Top 3 producing countries.
Top_1 =af_prod[af_prod['Country'] == 'Nigeria']
Top_2 =af_prod[af_prod['Country'] == 'Egypt']
Top_3 =af_prod[af_prod['Country'] == 'South Africa']
In [41]:
Top1_af_prod = Top_1.groupby(['Item'], as_index=False).Value.sum()
Top2_af_prod = Top_2.groupby(['Item'], as_index=False).Value.sum()
Top3_af_prod = Top_3.groupby(['Item'], as_index=False).Value.sum()
In [42]:
###Visualizing food items in Nigeria (Top 3 producing countries)
fig = px.bar(Top1_af_prod, 
             x ="Item", 
             y ="Value",
             color ='Value',
             hover_name ='Item',
             height=1000,
             title = 'Food Items Produced In Nigeria  (2004 - 2013)',
             labels = {'Year': 'Year', 'Value':'Value (kt)'})


fig.show()

Observation: Nigeria highest food produces are Cassava and its products & Yams.

In [43]:
###Visualizing food items in Egypt (Top 3 producing countries)
plt.figure(figsize=[12,7])
fig = px.bar(Top2_af_prod, 
             x ="Item", 
             y ="Value",
             color ='Value',
             hover_name ='Item',
             height=1000,
             title = 'Food Items Produced In Egypt  (2004 - 2013)',
             labels = {'Year': 'Year', 'Value':'Value (kt)'})


fig.show()

<Figure size 1200x700 with 0 Axes>

Observation: Egypt highest food produces is SugarCane although there is a relatively fair spread in their food production including Products like Maize & Products, Tomatoes,Vegetables, Wheat, Milk(excluding butter)unlike Nigeria.

In [44]:
###Visualizing food items in South Africa(Top 3 producing countries)
plt.figure(figsize=[12,7])
fig = px.bar(Top3_af_prod, 
             x ="Item", 
             y ="Value",
             color ='Value',
             hover_name ='Item',
             height=1000,
             title = 'Food Items Produced In South Africa  (2004 - 2013)',
             labels = {'Year': 'Year', 'Value':'Value (kt)'})


fig.show()

<Figure size 1200x700 with 0 Axes>

Observation: South Africa highest food produces is SugarCane and Maize.

QUESTION 8. Are there any outlier countries in the food production dataset between 2004 and 2013?¶

In [45]:
##Visualizing food supply data with seaborn boxplot
plt.figure(figsize=[12, 7])
base_color = sns.color_palette()[0]
sns.boxplot(data=af_prod, y='Value', x='Year', color=base_color)
plt.xlabel('Year')
plt.ylabel('Value (kt)')
plt.title('Food Produced Box Plot Per Year');
In [46]:
### Statistical plot showing means, quartiles, medians, etc.
af_prod.describe()
Out[46]:
Year Value
count 23110.000000 23110.000000
mean 2008.498269 327.785201
std 2.871740 1607.940343
min 2004.000000 0.000000
25% 2006.000000 3.000000
50% 2008.000000 18.000000
75% 2011.000000 108.000000
max 2013.000000 54000.000000
In [47]:
### Food Production interquatile range
af_prod_Q1=3.0
af_prod_Q3= 108.0
af_prod_IQR = af_prod_Q3 - af_prod_Q1
af_prod_IQR
Out[47]:
105.0
In [48]:
### Upper fence value
af_prod_upper_fence = af_prod_Q3 + ( 1.5 * af_prod_IQR)
af_prod_upper_fence
Out[48]:
265.5
In [49]:
### Lower fence value
af_prod_lower_fence = af_prod_Q1 + ( 1.5 * af_prod_IQR)
af_prod_lower_fence
Out[49]:
160.5

Finding out the outlier in food production -Any Country that has a value greater than the upper fence value is an outlier.¶

In [50]:
### Finding the countries whose value is higher than the upper fence
af_prod[af_prod.Value > (af_prod_Q1 + ( 1.5 * af_prod_IQR))]
Out[50]:
Country Item Year Value Unit
0 Algeria Wheat and products 2004 2731 kt
1 Algeria Wheat and products 2005 2415 kt
2 Algeria Wheat and products 2006 2688 kt
3 Algeria Wheat and products 2007 2319 kt
4 Algeria Wheat and products 2008 1111 kt
... ... ... ... ... ...
23085 Zimbabwe Milk - Excluding Butter 2009 389 kt
23086 Zimbabwe Milk - Excluding Butter 2010 396 kt
23087 Zimbabwe Milk - Excluding Butter 2011 400 kt
23088 Zimbabwe Milk - Excluding Butter 2012 410 kt
23089 Zimbabwe Milk - Excluding Butter 2013 410 kt

4830 rows × 5 columns

In [51]:
### Converting the outlier countries into a dataframe
out_countries = pd.DataFrame(af_prod[af_prod.Value > (af_prod_Q1 + ( 1.5 * af_prod_IQR))])
In [52]:
# checking the number of countries that are outliers
out_countries['Country'].nunique()
Out[52]:
39
In [53]:
# checking the unique names of the outlier countries
out_countries['Country'].unique()
Out[53]:
array(['Algeria', 'Angola', 'Benin', 'Burkina Faso', 'Cameroon',
       'Central African Republic', 'Chad', 'Congo', "Cote d'Ivoire",
       'Egypt', 'Ethiopia', 'Gabon', 'Ghana', 'Guinea', 'Guinea-Bissau',
       'Kenya', 'Liberia', 'Madagascar', 'Malawi', 'Mali', 'Mauritania',
       'Mauritius', 'Morocco', 'Mozambique', 'Namibia', 'Niger',
       'Nigeria', 'Rwanda', 'Senegal', 'Sierra Leone', 'South Africa',
       'Sudan', 'Swaziland', 'Togo', 'Tunisia', 'Uganda',
       'United Republic of Tanzania', 'Zambia', 'Zimbabwe'], dtype=object)
In [54]:
# Visualizing the country supply outlier with matplotlib (lineplot)
plt.figure(figsize=[12,7])
plt.plot(out_countries['Country'],out_countries['Value'],

    linewidth=1.5,
    linestyle='--',
    color= sns.color_palette()[6],
    marker='o',
    markersize=10,
    markerfacecolor=(0, 1, 1, 0.9))

plt.xticks(rotation=90)
plt.title('Food Producing Countries with Outliers')
plt.ylabel('Value (kt)')
plt.xlabel('Country');

Observation:39 countries in total produced consistently high amount of food between 2004 and 2013 and they are regarded as outlier countries. This Country include (Algeria, Angola, Benin, Burkina Faso, Cameroon, Central African Republic, Chad, Congo, Cote d'Ivoire, Egypt, Ethiopia, Gabon, Ghana, Guinea, Guinea-Bissau, Kenya, Liberia, Madagascar, Malawi, Mali, Mauritania, Mauritius, Morocco, Mozambique, Namibia, Niger, Nigeria, Rwanda, Senegal, Sierra Leone, South Africa, Sudan, Swaziland, Togo, Tunisia, Uganda, United Republic of Tanzania, Zambia, and Zimbabwe).

QUESTION 9. Are there any outlier countries in the food Supply dataset between 2004 and 2013?¶

In [55]:
##Visualizing food supply data with seaborn boxplot
plt.figure(figsize=[12, 7])
base_color = sns.color_palette()[5]
sns.boxplot(data=af_supply, y='Value', x='Year', color=base_color)
plt.xlabel('Year')
plt.ylabel('Value (kcal/person/day)')
plt.title('Food Supply Box Plot Per Year');

Observation It can be observed from the box plot that there exists a number of outliers in the food supply dataset for all the years from 2004-2013, with 2012 having the largest outlier. The dataset for the year 2012 will be explored to identify the country with the highest outliers.

In [56]:
#### Food supply data for 2012 only
af_supply_2012 = af_supply[af_supply['Year']==2012]
af_supply_2012
Out[56]:
Country Year Value Unit
8 Algeria 2012 3272 kcal/person/day
18 Angola 2012 2384 kcal/person/day
28 Benin 2012 2610 kcal/person/day
38 Botswana 2012 2342 kcal/person/day
48 Burkina Faso 2012 2707 kcal/person/day
58 Cabo Verde 2012 2590 kcal/person/day
68 Cameroon 2012 2622 kcal/person/day
78 Central African Republic 2012 2116 kcal/person/day
88 Chad 2012 2124 kcal/person/day
98 Congo 2012 2158 kcal/person/day
108 Cote d'Ivoire 2012 2780 kcal/person/day
118 Djibouti 2012 2545 kcal/person/day
128 Egypt 2012 3561 kcal/person/day
138 Ethiopia 2012 2124 kcal/person/day
148 Gabon 2012 2795 kcal/person/day
158 Gambia 2012 2652 kcal/person/day
168 Ghana 2012 3043 kcal/person/day
178 Guinea 2012 2561 kcal/person/day
188 Guinea-Bissau 2012 2331 kcal/person/day
198 Kenya 2012 2192 kcal/person/day
208 Lesotho 2012 2547 kcal/person/day
218 Liberia 2012 2200 kcal/person/day
228 Madagascar 2012 2058 kcal/person/day
238 Malawi 2012 2333 kcal/person/day
248 Mali 2012 2860 kcal/person/day
258 Mauritania 2012 2873 kcal/person/day
268 Mauritius 2012 3014 kcal/person/day
278 Morocco 2012 3366 kcal/person/day
288 Mozambique 2012 2298 kcal/person/day
298 Namibia 2012 2145 kcal/person/day
308 Niger 2012 2567 kcal/person/day
318 Nigeria 2012 2665 kcal/person/day
328 Rwanda 2012 2270 kcal/person/day
338 Sao Tome and Principe 2012 2338 kcal/person/day
348 Senegal 2012 2443 kcal/person/day
358 Sierra Leone 2012 2374 kcal/person/day
368 South Africa 2012 3014 kcal/person/day
370 Sudan 2012 2301 kcal/person/day
388 Swaziland 2012 2293 kcal/person/day
398 Togo 2012 2414 kcal/person/day
408 Tunisia 2012 3390 kcal/person/day
418 Uganda 2012 2160 kcal/person/day
428 United Republic of Tanzania 2012 2192 kcal/person/day
438 Zambia 2012 1923 kcal/person/day
448 Zimbabwe 2012 2197 kcal/person/day
In [57]:
#### Statistical description of 2012 data
af_supply.query('Year==2012')['Value'].describe()
Out[57]:
count      45.000000
mean     2527.644444
std       388.466926
min      1923.000000
25%      2200.000000
50%      2414.000000
75%      2707.000000
max      3561.000000
Name: Value, dtype: float64
In [58]:
### Food Supply interquatile range
af_supply_Q1=2200.000000
af_supply_Q3=2707.000000
af_supply_IQR = af_supply_Q3 - af_supply_Q1
af_supply_IQR
Out[58]:
507.0
In [59]:
### Upper fence value
af_supply_upper_fence = af_supply_Q3 + ( 1.5 * af_supply_IQR)
af_supply_upper_fence
Out[59]:
3467.5
In [60]:
### Lower fence value
af_supply_lower_fence= af_supply_Q1 + ( 1.5 * af_supply_IQR)
af_supply_lower_fence
Out[60]:
2960.5

Finding out the outlier in food Suppy -Any Country that has a value greater than the upper fence value is an outlier.¶

In [61]:
### Checking if the value of any country is lower than the lower fence value
af_supply_2012[af_supply.Value < (af_supply_Q1 - ( 1.5 * af_supply_IQR))].any()
Out[61]:
Country    False
Year       False
Value      False
Unit       False
dtype: bool
In [62]:
### Checking if the value of any country is higher than the upper fence
af_supply_2012[af_supply.Value < (af_supply_Q1 + ( 1.5 * af_supply_IQR))].any()
Out[62]:
Country    True
Year       True
Value      True
Unit       True
dtype: bool
In [63]:
### Finding the countries whose value is higher than the lower fence
af_supply_2012[af_supply.Value > (af_supply_Q1 + ( 1.5 * af_supply_IQR))]
Out[63]:
Country Year Value Unit
8 Algeria 2012 3272 kcal/person/day
128 Egypt 2012 3561 kcal/person/day
168 Ghana 2012 3043 kcal/person/day
268 Mauritius 2012 3014 kcal/person/day
278 Morocco 2012 3366 kcal/person/day
368 South Africa 2012 3014 kcal/person/day
408 Tunisia 2012 3390 kcal/person/day
In [64]:
### Checking if Egypt is the main outlier in the Supply dataset by year.
af_supply[af_supply.Value > (af_supply_Q3 + ( 1.5 * af_supply_IQR))]
Out[64]:
Country Year Value Unit
124 Egypt 2008 3490 kcal/person/day
126 Egypt 2010 3507 kcal/person/day
127 Egypt 2011 3549 kcal/person/day
128 Egypt 2012 3561 kcal/person/day
129 Egypt 2013 3522 kcal/person/day
In [65]:
# Visualizing the country supply outlier with matplotlib (lineplot)
plt.figure(figsize=[12, 7])
plt.plot(af_supply_2012['Country'],af_supply_2012['Value'],
    linewidth=1.5,
    linestyle='-',
    color='g',
    marker='o',
    markersize=10,
    markerfacecolor=(1, 0, 1, 0.5))

plt.xticks(rotation=90)
plt.title('Food Supplying Countries with Outlier in 2012');
plt.ylabel('Value (kcal/person/day)')
plt.xlabel('Country');

Observation:Egypt is the African country with highest food supply outliers because the supply value is higher than the upper fence value.It is also evident from the Visualization that Egypt has a substantially bigger food supply accessible for consumption in 2012 compared to other African countries.

QUESTION 10. Is there any relationship between Average Food Production and Food Supply per year?¶

In [66]:
##Remaining the production and supply dataframe
df1=af_prod
df2=af_supply
In [67]:
# The mean values of Food Production per Year
df1_mean = df1.groupby('Year', as_index=False).Value.mean()
df1_mean
Out[67]:
Year Value
0 2004 286.767301
1 2005 298.986592
2 2006 310.814014
3 2007 305.215830
4 2008 318.686851
5 2009 323.040657
6 2010 340.166955
7 2011 351.303633
8 2012 364.831816
9 2013 378.227568
In [68]:
#The mean values of Food Supplied per Year
df2_mean = df2.groupby('Year', as_index=False).Value.mean()
df2_mean
Out[68]:
Year Value
0 2004 2394.222222
1 2005 2409.288889
2 2006 2430.800000
3 2007 2447.755556
4 2008 2460.755556
5 2009 2482.222222
6 2010 2497.400000
7 2011 2515.422222
8 2012 2527.644444
9 2013 2532.244444
In [109]:
### Visualizing Average food produced and supplied between 2004-2013
plt.figure(figsize=[12, 7])
fig = make_subplots(rows=2, cols=1, shared_xaxes=True, vertical_spacing=0.02)

fig.add_trace(go.Scatter(x=df1_mean['Year'], y=df1_mean['Value'], name='Average Food Production'),
              row=2, col=1)

fig.add_trace(go.Scatter(x=df2_mean['Year'], y=df2_mean['Value'], name='Average Food Supply'),
              row=1, col=1)

fig.update_layout(height=800, width=1000, title_text="Average Food Production & Supply between 2004-2013");

fig.show()

<Figure size 1200x700 with 0 Axes>

Observation In 2007, the average food production dropped to 305.2158kt and the average food supplied was 2447.75kcal/person/day this means that more food was supplied to meet the need of the population.This situation highlights the fact that food production alone does not determine the availability of food for a population.There is a need for African Countries to improve their food productions capacities and reduce dependence on external sources of food.

QUESTION 11. Which African countries belong to the Top 5 OATS Producers?¶

In [70]:
###Selecting Oats production data
Oats_df = df1[df1.Item == 'Oats']
Oats_df
Out[70]:
Country Item Year Value Unit
40 Algeria Oats 2004 89 kt
41 Algeria Oats 2005 78 kt
42 Algeria Oats 2006 89 kt
43 Algeria Oats 2007 92 kt
44 Algeria Oats 2008 27 kt
... ... ... ... ... ...
22515 Zimbabwe Oats 2009 1 kt
22516 Zimbabwe Oats 2010 1 kt
22517 Zimbabwe Oats 2011 1 kt
22518 Zimbabwe Oats 2012 1 kt
22519 Zimbabwe Oats 2013 1 kt

90 rows × 5 columns

In [71]:
### Statistics showing the top 5 countries that produced Cassava
Oats_df.groupby(['Country'])['Value'].sum().sort_values(ascending=False).reset_index()[:5]
Out[71]:
Country Value
0 Algeria 849
1 Ethiopia 443
2 South Africa 406
3 Morocco 332
4 Kenya 37
In [72]:
###Visualing the Top 5 Oats Producing Countries (2004-2013)
Oats_producer = pd.DataFrame(Oats_df.groupby(['Country'])['Value'].sum().sort_values(ascending=False)).reset_index()[:5]
plt.figure(figsize=[12, 7])
ax = sns.barplot(y=Oats_producer.Value, x=Oats_producer.Country, data=Oats_df, color=None)
ax.set_title("Top 5 Oats Producing Countries (2004 - 2013)")
plt.xlabel('Country')
plt.ylabel('Value (kiloton)')

for i, bar in enumerate(ax.patches):
    count = bar.get_height()
    ax.annotate(f'{int(count)}',
                xy=(bar.get_x() + bar.get_width() / 2, bar.get_height()),
                ha='center', va='bottom', fontsize=13, color='black', fontweight='bold')

plt.show()

*Observation:The Top 5 Cassava Producing Countries that has the highest total quantity of Oats from 2004 to 2013 are: Algeria, Ethiopia, South Africa, Morrocco, and Kenya. However, Algeria is the top leading producer of Oat products.

QUESTION 12. Which African countries belong to the Top 5 Cassava Producers?¶

In [73]:
# Selecting Cassava production data
Cas_df = df1[df1.Item == 'Cassava and products']
Cas_df
Out[73]:
Country Item Year Value Unit
630 Angola Cassava and products 2004 8587 kt
631 Angola Cassava and products 2005 8806 kt
632 Angola Cassava and products 2006 9037 kt
633 Angola Cassava and products 2007 9730 kt
634 Angola Cassava and products 2008 10057 kt
... ... ... ... ... ...
22555 Zimbabwe Cassava and products 2009 216 kt
22556 Zimbabwe Cassava and products 2010 204 kt
22557 Zimbabwe Cassava and products 2011 225 kt
22558 Zimbabwe Cassava and products 2012 228 kt
22559 Zimbabwe Cassava and products 2013 230 kt

330 rows × 5 columns

In [74]:
### Statistics showing the top 5 countries that produced Cassava
Cas_df.groupby(['Country'])['Value'].sum().sort_values(ascending=False).reset_index()[:5]
Out[74]:
Country Value
0 Nigeria 452881
1 Ghana 121026
2 Angola 114286
3 Mozambique 71243
4 United Republic of Tanzania 52057
In [75]:
###Visualing the Top 5 Cassava Producing Countries 
Cass_producer = pd.DataFrame(Cas_df.groupby(['Country'])['Value'].sum().sort_values(ascending=False)).reset_index()[:5]

plt.figure(figsize=[12, 7])
ax = sns.barplot(y=Cass_producer.Value, x=Cass_producer.Country, data=Cas_df, color=None)
ax.set_title("Top 5 Cassava Producing Countries (2004 - 2013)")
plt.xlabel('Country')
plt.ylabel('Value (kiloton)')


for i, bar in enumerate(ax.patches):
    count = bar.get_height()
    ax.annotate(f'{int(count)}',
                xy=(bar.get_x() + bar.get_width() / 2, bar.get_height()),
                ha='center', va='bottom', fontsize=14, color='black', fontweight='bold')

plt.show()

*Observation:The Top 5 Cassava Producing Countries that has the highest total quantity of Cassava from 2004 to 2013 are: Nigeria,Ghana, Angola,Mozambique and United Republic of Tanzania.However, Nigeria is the top leading producer of Cassava.

QUESTION 13. Which African Countries belong to the Top 5 Beverage Producers?¶

In [76]:
# Selecting Beverage production data
Bev_df = df1[df1.Item == 'Beverages, Fermented']
Bev_df
Out[76]:
Country Item Year Value Unit
930 Angola Beverages, Fermented 2004 95 kt
931 Angola Beverages, Fermented 2005 99 kt
932 Angola Beverages, Fermented 2006 119 kt
933 Angola Beverages, Fermented 2007 131 kt
934 Angola Beverages, Fermented 2008 190 kt
... ... ... ... ... ...
22955 Zimbabwe Beverages, Fermented 2009 125 kt
22956 Zimbabwe Beverages, Fermented 2010 128 kt
22957 Zimbabwe Beverages, Fermented 2011 133 kt
22958 Zimbabwe Beverages, Fermented 2012 139 kt
22959 Zimbabwe Beverages, Fermented 2013 144 kt

300 rows × 5 columns

In [77]:
### Statistics showing the top 5 countries that produced Beverages
Bev_df.groupby(['Country'])['Value'].sum().sort_values(ascending=False).reset_index()[:5]
Out[77]:
Country Value
0 Nigeria 93839
1 Burkina Faso 30404
2 Uganda 29559
3 United Republic of Tanzania 25285
4 Cameroon 9829
In [78]:
###Visualing the Top 5 Beverage Producing Countries 
bev_producer = pd.DataFrame(Bev_df.groupby(['Country'])['Value'].sum().sort_values(ascending=False)).reset_index()[:5]
plt.figure(figsize=[12, 7])
ax = sns.barplot(y=bev_producer.Value, x=bev_producer.Country, data=Bev_df, color=None)
ax.set_title("Top 5 Beverage Producing Countries (2004 - 2013)", size=18)
plt.xlabel('Country', size=15)
plt.ylabel('Value (kiloton)', size=15)


for i, bar in enumerate(ax.patches):
    count = bar.get_height()
    ax.annotate(f'{int(count)}',
                xy=(bar.get_x() + bar.get_width() / 2, bar.get_height()),
                ha='center', va='bottom', fontsize=13, color='black', fontweight='bold')

plt.show()

*Observation:The Top 5 Producing Countries that has the highest total quantity of Beverage from 2004 to 2013 are: Nigeria, Burkina Faso, Uganda, United Republic of Tanzania, and Cameroon. However, Nigeria is the top leading producer of beverage.

QUESTION 14. Which African Countries belong to the Top 5 Rice Producers?¶

In [79]:
# Selecting Rice production data
Rice_df = df1[df1.Item =='Rice (Milled Equivalent)']
Rice_df
Out[79]:
Country Item Year Value Unit
10 Algeria Rice (Milled Equivalent) 2004 0 kt
11 Algeria Rice (Milled Equivalent) 2005 0 kt
12 Algeria Rice (Milled Equivalent) 2006 0 kt
13 Algeria Rice (Milled Equivalent) 2007 0 kt
14 Algeria Rice (Milled Equivalent) 2008 0 kt
... ... ... ... ... ...
22485 Zimbabwe Rice (Milled Equivalent) 2009 0 kt
22486 Zimbabwe Rice (Milled Equivalent) 2010 0 kt
22487 Zimbabwe Rice (Milled Equivalent) 2011 0 kt
22488 Zimbabwe Rice (Milled Equivalent) 2012 0 kt
22489 Zimbabwe Rice (Milled Equivalent) 2013 0 kt

380 rows × 5 columns

In [80]:
### Statistics showing the top 5 countries that produced Beverages
Rice_df.groupby(['Country'])['Value'].sum().sort_values(ascending=False).reset_index()[:5]
Out[80]:
Country Value
0 Egypt 40370
1 Nigeria 26964
2 Madagascar 26118
3 United Republic of Tanzania 10956
4 Guinea 10400
In [81]:
###Visualing the Top 5 Beverage Producing Countries 
Rice_producer = pd.DataFrame(Rice_df.groupby(['Country'])['Value'].sum().sort_values(ascending=False)).reset_index()[:5]
plt.figure(figsize=[12, 7])
ax = sns.barplot(y=Rice_producer.Value, x=Rice_producer.Country, data=Rice_df, color=None)
ax.set_title("TOP 5 Rice Producing Countries (2004 - 2013)", size=18)
plt.xlabel('Country', size=15)
plt.ylabel('Value (kiloton)', size=15)


for i, bar in enumerate(ax.patches):
    count = bar.get_height()
    ax.annotate(f'{int(count)}',
                xy=(bar.get_x() + bar.get_width() / 2, bar.get_height()),
                ha='center', va='bottom', fontsize=13, color='black', fontweight='bold')

plt.show()
In [82]:
### Visualizing top 5 African Producers of Rice (Milled Equivalent using a Map)
data = dict(
    type='choropleth',
    colorscale='viridis',
    locations=Rice_df['Country'],
    locationmode="country names",
    z=Rice_df['Value'],
    text=Rice_df['Country'],
    colorbar={'title': 'Rice Produced in Kiloton'},
)

layout = dict(
    title='Rice Produced in Africa (2004-2013)',
    geo=dict(
        scope='africa',
        showframe=False,
        projection={'type': 'mercator'}
    )
)

choromap = go.Figure(data=[data], layout=layout)

# Save the HTML file
pyo.plot(choromap, filename='rice_production_map.html')
Out[82]:
'rice_production_map.html'

Observation The Top 5 countries that have produced the highest total quantity of rice from 2004 to 2013 are:Egypt, Nigeria,Madagascar, United Republic of Tanzania and Guinea.However, Egypt is the top leading producer of Rice. This is a link to the interactive map- file:///C:/Users/aroko/rice_production_map.html. that shows the Top 5 Rice Producing Countries.

TEST OF HYPOTHESIS¶

Population growth may have an impact on Africa's high food production between 2004 and 2013.According to the FAO, rice was the third most produced food in the world in 2014.Rice (Milled Equivalent), a staple meal in most of the African Countries, may be used in hypothesis testing to gauge this.

The World Bank population dataset, the Country Code dataset, and the Rice (Milled Equivalent) production dataset for African Countries from 2004 to 2013 will be combined to provide test data for hypothesis measurement.

To analyze the linear relationship between our variables, X and Y, I will utilize the Pearson Correlation Coefficient. X: The Annual Volume of rice produced in African Countries Y: The Average annual population of African Countries.

The hypothesis test statements: H0 <= 0(NULL HYPOTHESIS), H1 > 0(ALTERNATIVE HYPOTHESIS)

NULL HYPOTHESIS: No significant correlation between population average in African countries and rice production in a specific year. i.e. no relationship between the two variables.

ALTERNATIVE HYPOTHESIS: African countries population and rice production are significantly correlated. i.e there is a relationship between the two variables.

QUESTION 15:Is there a relationship between Africa's population and the total amount of rice produced in African countries?¶

In [83]:
# Importing the Country Code dataset 

country_codes = pd.read_csv('2014 World_ GDP_with_Country_Codes.csv')
country_codes.head()
Out[83]:
COUNTRY GDP (BILLIONS) CODE
0 Afghanistan 21.71 AFG
1 Albania 13.40 ALB
2 Algeria 227.80 DZA
3 American Samoa 0.75 ASM
4 Andorra 4.80 AND
In [84]:
# Dropping and renaming columns in the country code dataset to fit the column names in the food production dataset
country_codes.rename(columns ={'COUNTRY':'Country','CODE':'Country Code'},inplace=True)
In [85]:
### Drop unwanted column 
country_codes.drop('GDP (BILLIONS)', axis=1, inplace=True)
country_codes.head()
Out[85]:
Country Country Code
0 Afghanistan AFG
1 Albania ALB
2 Algeria DZA
3 American Samoa ASM
4 Andorra AND
In [86]:
# Merging the country code dataset with the food production dataset

df1_codes = pd.merge(df1, country_codes, on ='Country', how = 'inner')
df1_codes.head()
Out[86]:
Country Item Year Value Unit Country Code
0 Algeria Wheat and products 2004 2731 kt DZA
1 Algeria Wheat and products 2005 2415 kt DZA
2 Algeria Wheat and products 2006 2688 kt DZA
3 Algeria Wheat and products 2007 2319 kt DZA
4 Algeria Wheat and products 2008 1111 kt DZA
In [87]:
# Merging the country code dataset with the food supply dataset
supp_code = pd.merge(df2, country_codes, on ='Country', how = 'inner')
supp_code
Out[87]:
Country Year Value Unit Country Code
0 Algeria 2004 2987 kcal/person/day DZA
1 Algeria 2005 2958 kcal/person/day DZA
2 Algeria 2006 3047 kcal/person/day DZA
3 Algeria 2007 3041 kcal/person/day DZA
4 Algeria 2008 3048 kcal/person/day DZA
... ... ... ... ... ...
415 Zimbabwe 2009 2147 kcal/person/day ZWE
416 Zimbabwe 2010 2168 kcal/person/day ZWE
417 Zimbabwe 2011 2200 kcal/person/day ZWE
418 Zimbabwe 2012 2197 kcal/person/day ZWE
419 Zimbabwe 2013 2110 kcal/person/day ZWE

420 rows × 5 columns

In [88]:
# Importing the population dataset

pop_df = pd.read_csv("population data(1960-2019).csv")
pop_df
Out[88]:
Country Name Year Count
0 Aruba 1960 54211
1 Afghanistan 1960 8996973
2 Angola 1960 5454933
3 Albania 1960 1608800
4 Andorra 1960 13411
... ... ... ...
12590 Kosovo 2017 1830700
12591 Yemen, Rep. 2017 27834821
12592 South Africa 2017 57000451
12593 Zambia 2017 16853688
12594 Zimbabwe 2017 14236745

12595 rows × 3 columns

In [89]:
# Renaming the country column in the population data
pop_df.rename(columns = {"Country Name":"Country", "Count": "Population"}, inplace=True)
pop_df
Out[89]:
Country Year Population
0 Aruba 1960 54211
1 Afghanistan 1960 8996973
2 Angola 1960 5454933
3 Albania 1960 1608800
4 Andorra 1960 13411
... ... ... ...
12590 Kosovo 2017 1830700
12591 Yemen, Rep. 2017 27834821
12592 South Africa 2017 57000451
12593 Zambia 2017 16853688
12594 Zimbabwe 2017 14236745

12595 rows × 3 columns

In [90]:
country_codes.head()
Out[90]:
Country Country Code
0 Afghanistan AFG
1 Albania ALB
2 Algeria DZA
3 American Samoa ASM
4 Andorra AND
In [91]:
# Merging population dataset with the country code dataset
pop_df_code = pd.merge(pop_df, country_codes, on ='Country', how = 'inner')
In [92]:
pop_df_code['Country'].unique()
Out[92]:
array(['Aruba', 'Afghanistan', 'Angola', 'Albania', 'Andorra',
       'United Arab Emirates', 'Argentina', 'Armenia', 'American Samoa',
       'Antigua and Barbuda', 'Australia', 'Austria', 'Azerbaijan',
       'Burundi', 'Belgium', 'Benin', 'Burkina Faso', 'Bangladesh',
       'Bulgaria', 'Bahrain', 'Bahamas, The', 'Bosnia and Herzegovina',
       'Belarus', 'Belize', 'Bermuda', 'Bolivia', 'Brazil', 'Barbados',
       'Bhutan', 'Botswana', 'Central African Republic', 'Canada',
       'Switzerland', 'Chile', 'China', "Cote d'Ivoire", 'Cameroon',
       'Colombia', 'Comoros', 'Cabo Verde', 'Costa Rica', 'Cuba',
       'Curacao', 'Cayman Islands', 'Cyprus', 'Czech Republic', 'Germany',
       'Djibouti', 'Dominica', 'Denmark', 'Dominican Republic', 'Algeria',
       'Ecuador', 'Eritrea', 'Spain', 'Estonia', 'Ethiopia', 'Finland',
       'Fiji', 'France', 'Faroe Islands', 'Gabon', 'United Kingdom',
       'Georgia', 'Ghana', 'Gibraltar', 'Guinea', 'Gambia, The',
       'Guinea-Bissau', 'Equatorial Guinea', 'Greece', 'Grenada',
       'Greenland', 'Guatemala', 'Guam', 'Guyana', 'Honduras', 'Croatia',
       'Haiti', 'Hungary', 'Indonesia', 'Isle of Man', 'India', 'Ireland',
       'Iraq', 'Iceland', 'Israel', 'Italy', 'Jamaica', 'Jordan', 'Japan',
       'Kazakhstan', 'Kenya', 'Cambodia', 'Kiribati', 'Kuwait', 'Lebanon',
       'Liberia', 'Libya', 'Liechtenstein', 'Sri Lanka', 'Lesotho',
       'Lithuania', 'Luxembourg', 'Latvia', 'Morocco', 'Monaco',
       'Moldova', 'Madagascar', 'Maldives', 'Mexico', 'Marshall Islands',
       'Mali', 'Malta', 'Montenegro', 'Mongolia',
       'Northern Mariana Islands', 'Mozambique', 'Mauritania',
       'Mauritius', 'Malawi', 'Malaysia', 'Namibia', 'New Caledonia',
       'Niger', 'Nigeria', 'Nicaragua', 'Netherlands', 'Norway', 'Nepal',
       'New Zealand', 'Oman', 'Pakistan', 'Panama', 'Peru', 'Philippines',
       'Palau', 'Papua New Guinea', 'Poland', 'Puerto Rico', 'Portugal',
       'Paraguay', 'French Polynesia', 'Qatar', 'Romania', 'Rwanda',
       'Saudi Arabia', 'Sudan', 'Senegal', 'Singapore', 'Solomon Islands',
       'Sierra Leone', 'El Salvador', 'San Marino', 'Somalia',
       'South Sudan', 'Sao Tome and Principe', 'Suriname', 'Slovenia',
       'Sweden', 'Seychelles', 'Chad', 'Togo', 'Thailand', 'Tajikistan',
       'Turkmenistan', 'Timor-Leste', 'Tonga', 'Trinidad and Tobago',
       'Tunisia', 'Turkey', 'Tuvalu', 'Tanzania', 'Uganda', 'Ukraine',
       'Uruguay', 'United States', 'Uzbekistan', 'British Virgin Islands',
       'Vietnam', 'Vanuatu', 'Samoa', 'Kosovo', 'South Africa', 'Zambia',
       'Zimbabwe', 'Serbia'], dtype=object)
In [93]:
pop_df_code['Country'].nunique()
Out[93]:
187
In [94]:
# Creating a list to hold the 187 African countries in the production and supply dataset
af_pop_country_list = pop_df_code['Country'].unique()
af_pop_country_list
Out[94]:
array(['Aruba', 'Afghanistan', 'Angola', 'Albania', 'Andorra',
       'United Arab Emirates', 'Argentina', 'Armenia', 'American Samoa',
       'Antigua and Barbuda', 'Australia', 'Austria', 'Azerbaijan',
       'Burundi', 'Belgium', 'Benin', 'Burkina Faso', 'Bangladesh',
       'Bulgaria', 'Bahrain', 'Bahamas, The', 'Bosnia and Herzegovina',
       'Belarus', 'Belize', 'Bermuda', 'Bolivia', 'Brazil', 'Barbados',
       'Bhutan', 'Botswana', 'Central African Republic', 'Canada',
       'Switzerland', 'Chile', 'China', "Cote d'Ivoire", 'Cameroon',
       'Colombia', 'Comoros', 'Cabo Verde', 'Costa Rica', 'Cuba',
       'Curacao', 'Cayman Islands', 'Cyprus', 'Czech Republic', 'Germany',
       'Djibouti', 'Dominica', 'Denmark', 'Dominican Republic', 'Algeria',
       'Ecuador', 'Eritrea', 'Spain', 'Estonia', 'Ethiopia', 'Finland',
       'Fiji', 'France', 'Faroe Islands', 'Gabon', 'United Kingdom',
       'Georgia', 'Ghana', 'Gibraltar', 'Guinea', 'Gambia, The',
       'Guinea-Bissau', 'Equatorial Guinea', 'Greece', 'Grenada',
       'Greenland', 'Guatemala', 'Guam', 'Guyana', 'Honduras', 'Croatia',
       'Haiti', 'Hungary', 'Indonesia', 'Isle of Man', 'India', 'Ireland',
       'Iraq', 'Iceland', 'Israel', 'Italy', 'Jamaica', 'Jordan', 'Japan',
       'Kazakhstan', 'Kenya', 'Cambodia', 'Kiribati', 'Kuwait', 'Lebanon',
       'Liberia', 'Libya', 'Liechtenstein', 'Sri Lanka', 'Lesotho',
       'Lithuania', 'Luxembourg', 'Latvia', 'Morocco', 'Monaco',
       'Moldova', 'Madagascar', 'Maldives', 'Mexico', 'Marshall Islands',
       'Mali', 'Malta', 'Montenegro', 'Mongolia',
       'Northern Mariana Islands', 'Mozambique', 'Mauritania',
       'Mauritius', 'Malawi', 'Malaysia', 'Namibia', 'New Caledonia',
       'Niger', 'Nigeria', 'Nicaragua', 'Netherlands', 'Norway', 'Nepal',
       'New Zealand', 'Oman', 'Pakistan', 'Panama', 'Peru', 'Philippines',
       'Palau', 'Papua New Guinea', 'Poland', 'Puerto Rico', 'Portugal',
       'Paraguay', 'French Polynesia', 'Qatar', 'Romania', 'Rwanda',
       'Saudi Arabia', 'Sudan', 'Senegal', 'Singapore', 'Solomon Islands',
       'Sierra Leone', 'El Salvador', 'San Marino', 'Somalia',
       'South Sudan', 'Sao Tome and Principe', 'Suriname', 'Slovenia',
       'Sweden', 'Seychelles', 'Chad', 'Togo', 'Thailand', 'Tajikistan',
       'Turkmenistan', 'Timor-Leste', 'Tonga', 'Trinidad and Tobago',
       'Tunisia', 'Turkey', 'Tuvalu', 'Tanzania', 'Uganda', 'Ukraine',
       'Uruguay', 'United States', 'Uzbekistan', 'British Virgin Islands',
       'Vietnam', 'Vanuatu', 'Samoa', 'Kosovo', 'South Africa', 'Zambia',
       'Zimbabwe', 'Serbia'], dtype=object)
In [95]:
##  Checking the names of the African Countries from the food_production dataset
df1['Country'].unique()
Out[95]:
array(['Algeria', 'Angola', 'Benin', 'Botswana', 'Burkina Faso',
       'Cabo Verde', 'Cameroon', 'Central African Republic', 'Chad',
       'Congo', "Cote d'Ivoire", 'Djibouti', 'Egypt', 'Ethiopia', 'Gabon',
       'Gambia', 'Ghana', 'Guinea', 'Guinea-Bissau', 'Kenya', 'Lesotho',
       'Liberia', 'Madagascar', 'Malawi', 'Mali', 'Mauritania',
       'Mauritius', 'Morocco', 'Mozambique', 'Namibia', 'Niger',
       'Nigeria', 'Rwanda', 'Sao Tome and Principe', 'Senegal',
       'Sierra Leone', 'South Africa', 'Sudan', 'Swaziland', 'Togo',
       'Tunisia', 'Uganda', 'United Republic of Tanzania', 'Zambia',
       'Zimbabwe'], dtype=object)
In [96]:
df1['Country'].nunique()
Out[96]:
45
In [97]:
# Creating a list to hold the 45 African countries in the production and supply dataset
af_prod_country_list =['Algeria', 'Angola', 'Benin', 'Botswana', 'Burkina Faso',
       'Cabo Verde', 'Cameroon', 'Central African Republic', 'Chad',
       'Congo', "Cote d'Ivoire", 'Djibouti', 'Egypt', 'Ethiopia', 'Gabon',
       'Gambia', 'Ghana', 'Guinea', 'Guinea-Bissau', 'Kenya', 'Lesotho',
       'Liberia', 'Madagascar', 'Malawi', 'Mali', 'Mauritania',
       'Mauritius', 'Morocco', 'Mozambique', 'Namibia', 'Niger',
       'Nigeria', 'Rwanda', 'Sao Tome and Principe', 'Senegal',
       'Sierra Leone', 'South Africa', 'Sudan', 'Swaziland', 'Togo',
       'Tunisia', 'Uganda', 'United Republic of Tanzania', 'Zambia',
       'Zimbabwe']

The population dataset has 187 countries and the african_production and african supply datasets has 45 countries each.

In [98]:
##Dropping the population country data
pop_df_code.drop("Country", axis = 1, inplace = True)
pop_df_code
Out[98]:
Year Population Country Code
0 1960 54211 ABW
1 1961 55438 ABW
2 1962 56225 ABW
3 1963 56695 ABW
4 1964 57032 ABW
... ... ... ...
10802 2013 7164132 SRB
10803 2014 7130576 SRB
10804 2015 7095383 SRB
10805 2016 7058322 SRB
10806 2017 7020858 SRB

10807 rows × 3 columns

In [99]:
###Checking the population Year datatype
pop_df_code.dtypes
Out[99]:
Year             int64
Population       int64
Country Code    object
dtype: object
In [100]:
# Extracting rice production data in African countries over the years
Rice_df = df1.query("Item == 'Rice (Milled Equivalent)'")
Rice_total = pd.DataFrame(Rice_df.groupby(['Country', 'Year'], as_index = False)['Value'].sum())
Rice_total
Out[100]:
Country Year Value
0 Algeria 2004 0
1 Algeria 2005 0
2 Algeria 2006 0
3 Algeria 2007 0
4 Algeria 2008 0
... ... ... ...
375 Zimbabwe 2009 0
376 Zimbabwe 2010 0
377 Zimbabwe 2011 0
378 Zimbabwe 2012 0
379 Zimbabwe 2013 0

380 rows × 3 columns

In [101]:
# Merging the rice dataset with the country code dataset
rice_df = pd.merge(Rice_total, country_codes, on ='Country', how = 'inner')

# Renaming the Value column in the rice production data
rice_df.rename(columns = {"Value":"Rice Production (Kilotonnes)"}, inplace=True)
rice_df.head(4)
Out[101]:
Country Year Rice Production (Kilotonnes) Country Code
0 Algeria 2004 0 DZA
1 Algeria 2005 0 DZA
2 Algeria 2006 0 DZA
3 Algeria 2007 0 DZA
In [102]:
#Merging the population and rice production dataset using the country code and year
Rice_Pop_df = pd.merge(rice_df, pop_df_code, on= [ "Year"], how = "inner")
Rice_Pop_df.head()
Out[102]:
Country Year Rice Production (Kilotonnes) Country Code_x Population Country Code_y
0 Algeria 2004 0 DZA 98737 ABW
1 Algeria 2004 0 DZA 24726684 AFG
2 Algeria 2004 0 DZA 18758145 AGO
3 Algeria 2004 0 DZA 3026939 ALB
4 Algeria 2004 0 DZA 76244 AND
In [103]:
##Dropping the country code
Rice_Pop_df.drop("Country Code_x", axis = 1, inplace = True)
Rice_Pop_df.drop("Country Code_y", axis = 1, inplace = True)
Rice_Pop_df.head()
Out[103]:
Country Year Rice Production (Kilotonnes) Population
0 Algeria 2004 0 98737
1 Algeria 2004 0 24726684
2 Algeria 2004 0 18758145
3 Algeria 2004 0 3026939
4 Algeria 2004 0 76244
In [104]:
# Computing the average rice production and popopulation per year
Rice_production_df = pd.DataFrame(Rice_Pop_df.groupby("Year", as_index=False).mean())
Rice_production_df
Out[104]:
Year Rice Production (Kilotonnes) Population
0 2004 333.971429 3.129071e+07
1 2005 356.028571 3.168758e+07
2 2006 387.885714 3.208640e+07
3 2007 366.657143 3.248575e+07
4 2008 428.457143 3.288969e+07
5 2009 413.657143 3.329366e+07
6 2010 444.057143 3.369604e+07
7 2011 449.228571 3.409119e+07
8 2012 488.371429 3.466432e+07
9 2013 488.371429 3.507703e+07
In [105]:
# Performing the Pearson Correlation hypothesis to measure the correlation between the average rice production and population
from scipy.stats import pearsonr
x = np.array(Rice_production_df['Rice Production (Kilotonnes)'])
y = np.array(Rice_production_df['Population'])

pearsonr(x, y)
Out[105]:
PearsonRResult(statistic=0.9694043205583609, pvalue=3.694735823540323e-06)
In [106]:
##To Calculate the pearson corelation
import numpy
numpy.corrcoef(y, x)[0, 1]
Out[106]:
0.9694043205583608
In [107]:
pd.DataFrame(pearsonr(y,x),index= (['statistic', 'p value']),columns=['pearson correlation'])
Out[107]:
pearson correlation
statistic 0.969404
p value 0.000004
In [108]:
# let's visualize the correlation result with a scatter plot
plt.figure(figsize=[12, 7])
sns.regplot(data=Rice_production_df, x = "Rice Production (Kilotonnes)", y = "Population",  x_jitter=0.7, fit_reg=True, 
            scatter_kws={'alpha':0.9})

plt.title('Relationship between Population vs Rice Production (2004 - 2013)')
plt.ylabel('"Rice Production (Kt)')
plt.xlabel('Population');

Observation

The correlation hypothesis shows a significant positive relationship between rice production and African countries population during a specific year, the small p value shows that the null hypothesis (H0) is rejected at the 5% level.

The correlation coefficient "0.97" shows a strong relationship between average rice production each year and average African population size in the same year.

However, despite a significant relationship between population and rice production, this does not necessarily imply that rice production is what causes population growth.

INSIGHTS FROM THE ANALYSIS¶

1.Food production in Africa increases yearly. However, apart from the drop in 2007, there was a progress increase in food production from 2008 to 2013.

2.There is a progressive increase in food supply in African countries from 2004 to 2013.

3.The TOP 3 African countries with the highest volume of food production within (2004-2013) are Nigeria, Egypt, & South Africa. However, the BOTTOM 3 African countries Cabo Verde, Sao Tome & Principe and Djibouti produced the least volume of food.

4.The TOP 3 African countries with the highest volume of food supply within (2004-2013) are Egypt, Tunisia, & Morocco. However, the BOTTOM 3 African countries Chad, Ethiopia and Zambia supplied the least volume of food..

5.The Top 5 most produced food in Africa between 2004 and 2013 are Cassava, Sugar Cane, Maize and products, Yam, and Vegetables.

6.The Least 5 most produced food in Africa between 2004 and 2013 are Cloves, Molluscs,Pepper, Fish,Liver oil and Aquatic Animals.

7.Among the Top 3 African producing countries:

Nigeria highest food produces are Cassava and its products & Yams.The reason for this high production in these two categories could be attributed to the ease of cultivating and producing these crops in Nigeria.The combination of favorable climatic conditions, cultural preferences, and relatively straightforward cultivation methods contributes to Nigeria's high production of Cassava and Yams.

Egypt highest food produces is Sugarcane although there is a relatively fair spread in their food production including Products like Maize & Products, Tomatoes, Vegetables, Wheat, Milk (excluding butter) unlike Nigeria.

South Africa highest food produces is SugarCane and Maize.

8.39 countries in total produced consistently high amount of food between 2004 and 2013 and they are regarded as outlier countries. This Country include (Algeria, Angola, Benin, Burkina Faso, Cameroon, Central African Republic, Chad, Congo, Cote d'Ivoire, Egypt, Ethiopia, Gabon, Ghana, Guinea, Guinea-Bissau, Kenya, Liberia, Madagascar, Malawi, Mali, Mauritania, Mauritius, Morocco, Mozambique, Namibia, Niger, Nigeria, Rwanda, Senegal, Sierra Leone, South Africa, Sudan, Swaziland, Togo, Tunisia, Uganda, United Republic of Tanzania, Zambia, and Zimbabwe).

9.Egypt is the African country with highest food supply outliers because the supply value is higher than the upper fence value. It is also evident from the Visualization that Egypt has a substantially bigger food supply accessible for consumption in 2012 compared to other African countriess.

10.In 2007, the average food production dropped to 305.2158kt and the average food supplied was 2447.75kcal/person/day this means that more food was supplied to meet the need of the population. This situation highlights the fact that food production alone does not determine the availability of food for a population. There is a need for African Countries to improve their food productions capacities and reduce dependence on external sources of food.

11.The Top 5 Cassava Producing Countries that has the highest total quantity of Oats from 2004 to 2013 are: Algeria, Ethiopia, South Africa, Morrocco, and Kenya. However, Algeria is the top leading producer of Oat products.

12.The Top 5 Cassava Producing Countries that has the highest total quantity of Cassava from 2004 to 2013 are: Nigeria, Ghana, Angola, Mozambique, and United Republic of Tanzania. However, Nigeria is the top leading producer of Cassava.

13.The Top 5 Producing Countries that has the highest total quantity of Beverage from 2004 to 2013 are: Nigeria, Burkina Faso, Uganda, United Republic of Tanzania, and Cameroon. However, Nigeria is the top leading producer of beverage.

14.The Top 5 countries that have produced the highest total quantity of rice from 2004 to 2013 are: Egypt, Nigeria, Madagascar, United Republic of Tanzania, and Guinea. However, Egypt is the top leading producer of Rice.

15.The correlation hypothesis shows a significant positive relationship between rice production and African countries population during a specific year, the small p value shows that the null hypothesis (H0) is rejected at the 5% level. The correlation coefficient "0.97" shows a strong relationship between average rice production each year and average African population size in the same year. However, despite a significant relationship between population and rice production, correlation doesn't imply causation. This does not necessarily imply that rice production is what causes population growth.

Conclusion:¶

According to the Hypothesis testing, there was a bit significant relationship between population, food supply, and production in Africa between 2004 and 2013. Between 2004 and 2013, African countries experienced a potential food shortage due to a low food supply compared to increased production of food and population.

However, the population doesn't seem to explain the possible cause of the food shortage in Africa during this time other elements like price inflation, exports, and inadequate storage facilities may be reasons for the shortage because countries with high food production have smaller food supply compared to their production.

African countries must embark on a more deliberate transition from subsistence to commercial agriculture, this will enhance both the food supply and the countries total GDP. It will also help Africa country decrease or eradicate global hunger.

In [ ]: